On 29 July 2016, the European Banking Authority (‘EBA’) released the results of yet another round of stress tests for major European credit institutions. Of the 51 institutions participating in the exercise, Monte dei Paschi di Sienna (‘MPS’) –the world’s oldest and Italy’s third largest bank– fared worse, with its ratio of CET1 capital to risk-weighted assets falling to as low as -2.41% in the adverse scenario, which assumed a three-year-long severe recession. As a result, the ECB, acting in its capacity as direct supervisor of the euro area’s largest banks, requested MPS to proceed within a few months to full recapitalization, to an extent sufficient for the bank to meet its basic regulatory requirements even under the adverse scenario. Eight months down the line, this has not yet happened and MPS remains in limbo. The impasse is directly related to the bank resolution regime hastily adopted at EU level in the wake of the global financial crisis.
Role of the European anti-bailout norms
The resolution regime is premised on the belief that, depending on the circumstances, keeping certain failed or failing banks open for business may be necessary for reasons of systemic stability. At the same time, it is explicitly designed to put an end to state-funded bailouts. This fundamental policy choice is justified on two conceptually distinct grounds: the protection of taxpayers; and the containment of moral hazard and competitive distortions in the banking system.
The priority of bail-in over state-based assistance is reflected in the two distinct but interrelated sets of European norms governing bank resolution actions: the Commission’s norms on state aids in the banking sector, as set out in the Banking Communication of July 2013; and the new special resolution regime for credit institutions and investment firms adopted in May 2014 in the form of the Bank Recovery and Resolution Directive (BRRD). The two regimes run in parallel and their combined effect is to render outright bailouts almost impossible.
As I explain in detail in a recent paper on ‘Limits on State-Funded Bailouts in the EU Bank Resolution Regime’, the BRRD permits the injection of state funds in a distressed bank on grounds of systemic safety. However, there are strict preconditions, and the intervention can only take place in strictly circumscribed forms. What is more significant, the public intervention may only occur at a late stage within the BRRD’s resolution financing cascade – that is, after substantial bail-in of the claims of the ailing banks’ junior, and even certain classes of senior, private debtholders, amounting to at least 8% of total liabilities before the state can participate in the banks’ rescue. The only exception under the BRRD is a so-called ‘precautionary recapitalization’. This involves the injection of state funds in institutions which are still solvent, in the sense that they still meet their basic regulatory capital requirements. Significantly, precautionary recapitalization cannot be used to absorb past losses, which must be covered out of private resources prior to the state’s intervention. Furthermore, the European Commission may require prior burden-sharing by private stakeholders (that is, bail-in by any other name, albeit of more modest proportions) as a condition for approving the precautionary recapitalization under the state-aid regime.
This, to be sure, has been the main impediment to a rapid recapitalization of MPS. With a hoped-for private-sector solution failing to materialize, the Italian government started pushing for a publicly funded recapitalization.[] However, it found the European Commission unbending in its insistence on ‘burden-sharing’ by subordinated debtholders. This has led to a protracted tripartite process, with the Italian government seeking a solution agreeable to both the Commission and the ECB.
Ambiguities of the European policy
The case brings to light certain ambiguities and potential shortcomings of the European policy.
- To start with, the inordinate length of the interactions between the Commission, the ECB and the national government is in itself a destabilizing factor. Ideally, a weak bank’s predicament should be addressed within the proverbial weekend. MPS’s bailout is under discussion since last autumn and the result is not yet clear. In the meantime, the bank’s liquidity position has deteriorated badly, forcing the Italian government, with the Commission’s connivance, to provide ample liquidity support. Insofar as liquidity support is just another form of publicly funded rescue operation, this raises questions about the coherence of the Commission’s policy stance.
- Bailing in subordinated debt instruments is a political problem for the Italian government, because these were primarily sold (probably without sufficient disclosure) to unsuspecting retail investors, rather than financial institutions. While the government appears to accept that a conversion of subordinated debt into equity cannot be avoided, the intention is to offer some form of compensation. But this entails its own legal difficulties. This facet of the case shows the difficulties involved in applying retrospectively the concept of bail-in to old debt instruments.
- In view of the weak state of the Italian banking market, bailing in MPS’s debtholders also raises the possibility of contagion. In a weak environment and in a context of widespread distress, bail-in of one bank’s claimholders may lead those of other banks to reappraise their position, precipitating an across-the-board flight to quality. Empirically, it must be noted that the tough approach of the ECB and the Commission has not caused a panic, as many had originally feared. Instead, the market has differentiated between the liabilities of MPS and other weak banks and those of their better capitalized peers, like Intesa Sanpaolo. Moreover, Italy’s largest bank, Unicredit, has now completed successfully a huge voluntary recapitalization, raising €13 billion in new equity. This may be interpreted as proof that the stringency of the supervisory approach has succeeded in credibly signaling the various banks’ relative strength.
- In relation to the country’s weaker banks, however, the supervisory insistence on the immediate and comprehensive restoration of capital positions may have made the implementation of private solutions more difficult. Any investor considering to participate in a capital-raising plan, or simply to lend them money, will now think twice, knowing that, if for one reason or another the bank’s private funding proves to be less than fully sufficient, bail in will ensue without respite.
Legal conditions for precautionary recapitalization
We should recall that the BRRD’s provisions only allow the state to proceed with a precautionary recapitalization when the bank in question meets its basic capital requirements. Moreover, the state’s capital injection cannot be used to offset past losses, but only to cover a capital shortfall identified as a result of a national or pan-European stress test. In the case of the MPS, it was accepted that the bank qualifies for precautionary recapitalization; but the Commission was adamant that it should not be exempted from burden-sharing.
This raises two further issues:
- The estimation of the capital shortfall and, in particular, of the extent to which it is attributable to past losses is by no means an exact science. This is especially true of expected but as yet unrealized losses on existing assets, including the going value of portfolios of NPLs. Much depends on assumptions about macroeconomic dynamics and the time horizon within which these should be disposed of or run off. Due to the valuations’ wide margin of error, the supposedly rigid norm becomes malleable, and the resolution system can either be applied in an extremely harsh way or be ‘contaminated’ by supervisory forbearance – but this depends on contingent and largely discretionary appraisals.
- A precautionary recapitalization is only permissible if a bank is still solvent; by necessary implication, demanding burden-sharing in this context is tantamount to subjecting junior debtholders to losses at a clearly pre-insolvency stage. This may be advisable as a matter of policy, but not necessary as a matter of principle. One wonders, accordingly, whether the Commission’s policy of strict insistence on bail-in or burden-sharing in all cases of perceived undercapitalization is wise.
An alternative path?
Most commentators see the MPS case solely as a test for the credibility of the European resolution regime, which will apparently be compromised, if the rules are bent to permit the Italian state to recapitalize MPS without applying bail-in. Nevertheless, this is to forget the main rationale for the establishment of a special bank resolution regime, separate from the general insolvency framework, that is, the preservation of systemic stability. In conditions of economic distress and system-wide banking weakness, bail-in as the preferred, and essentially mandatory, resolution tool can aggravate the situation (see, eg, De Grawe, 2013, and Persaud, 2014).
Arguably, the current approach places undue emphasis on bank-level capital positions. The Italian situation is characterized by a number of medium-sized weak banks, an enormous level of bad assets (with some €360bn of NPEs & doubtful debts as of late 2016), no clear path to NPL resolution (also due to missing or incomplete secondary markets), and a weak economy. In these circumstances, a system-wide pre-resolution action plan, involving the management of impaired assets within a realistic timeframe, as well as the recapitalization of weak banks with state aid but without extensive burden-sharing, may constitute the most credible and reasonable response. While rather difficult to implement under the existing European legal framework, an approach of this type might yield better results, both in terms of delivering systemic stability and of protecting taxpayers.
Christos Hadjiemmanuil is Professor of Monetary and Financial Institutions at the University of Piraeus and Visiting Professor of Law at the London School of Economics and Political Science.